Multi-user detection allows for the efficient use of band-width in code-division multiple-access (CDMA) channels through mitigation of near-far effects and multiple-access noise limitations. Due to its inherent noise and multi-path immunity, CDMA multi-access is being considered as a platform for personal communication systems (PCS). As CDMA based digital communication networks proliferate, the need to determine the presence of a new user and integrate knowledge of this new user into the detection scheme becomes more important. The decorrelating detector is a linear multi-user detector that is asymptotically optimal in terms of near far resistance; however, in the presence of a new unknown user, performance of the decorrelator is severely degraded. Adaptive decorrelators are constructed which adaptively augment an existing conventional decorrelator to demodulate a new active user in addition to existing risers. Several likelihood ratio based schemes are proposed.
{"title":"Adaptive decorrelating detectors for CDMA channels","authors":"U. Mitra, H. Poor","doi":"10.1109/ICC.1995.524266","DOIUrl":"https://doi.org/10.1109/ICC.1995.524266","url":null,"abstract":"Multi-user detection allows for the efficient use of band-width in code-division multiple-access (CDMA) channels through mitigation of near-far effects and multiple-access noise limitations. Due to its inherent noise and multi-path immunity, CDMA multi-access is being considered as a platform for personal communication systems (PCS). As CDMA based digital communication networks proliferate, the need to determine the presence of a new user and integrate knowledge of this new user into the detection scheme becomes more important. The decorrelating detector is a linear multi-user detector that is asymptotically optimal in terms of near far resistance; however, in the presence of a new unknown user, performance of the decorrelator is severely degraded. Adaptive decorrelators are constructed which adaptively augment an existing conventional decorrelator to demodulate a new active user in addition to existing risers. Several likelihood ratio based schemes are proposed.","PeriodicalId":241383,"journal":{"name":"Proceedings IEEE International Conference on Communications ICC '95","volume":"80 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1995-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133504977","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The leaky bucket (LB) is emerging as the best possible candidate for implementing the policing mechanism in ATM networks. However, it exhibits a strong coupling between two conflicting requirements, viz., accurate average rate enforcement and prompt reaction to sudden bursts. We propose a modification to the LB mechanism which, while providing a more accurate average rate enforcement (closer to ideal "policer" behavior), also reacts faster (i.e., detects violating cells) to prevent long bursts. The additional hardware and complexity introduced by this modification is marginal. Briefly, the mechanism can be described as follows: every second cell arriving from a source is monitored by a LB while a second LB "polices" the alternate cell stream. This has the effect of reducing the squared coefficient of variation of interarrival times by a factor of two. The predictability of the arrival stream is increased with this mechanism and consequently, the parameters of the LB can be dimensioned more precisely, resulting in a more effective mechanism.
{"title":"The effect of variance reduction on the performance of the leaky bucket","authors":"R. Krishnan, J. Silvester","doi":"10.1109/ICC.1995.524540","DOIUrl":"https://doi.org/10.1109/ICC.1995.524540","url":null,"abstract":"The leaky bucket (LB) is emerging as the best possible candidate for implementing the policing mechanism in ATM networks. However, it exhibits a strong coupling between two conflicting requirements, viz., accurate average rate enforcement and prompt reaction to sudden bursts. We propose a modification to the LB mechanism which, while providing a more accurate average rate enforcement (closer to ideal \"policer\" behavior), also reacts faster (i.e., detects violating cells) to prevent long bursts. The additional hardware and complexity introduced by this modification is marginal. Briefly, the mechanism can be described as follows: every second cell arriving from a source is monitored by a LB while a second LB \"polices\" the alternate cell stream. This has the effect of reducing the squared coefficient of variation of interarrival times by a factor of two. The predictability of the arrival stream is increased with this mechanism and consequently, the parameters of the LB can be dimensioned more precisely, resulting in a more effective mechanism.","PeriodicalId":241383,"journal":{"name":"Proceedings IEEE International Conference on Communications ICC '95","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1995-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133709876","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The pre-arbitrated scheme in the DQDB network is suitable for constant-rate and connection-oriented transmission. The authors propose a periodic assignment scheme, which classifies isochronous connections into several harmonic classes according to their periods. The main advantage of the proposed scheme is that by assigning the same unit channel to connections of the same class, these connections can be established as long as the total required bandwidth is no greater than the available bandwidth of the unit channel. The authors study three assignment policies based on the harmonic case of the pinwheel algorithm. These assignment policies are evaluated in terms of their blocking probabilities through both simulation and analysis. It is found that the busy-first policy, which tries to assign a new isochronous connection with a unit channel already carrying other connections of the same class, is the most suitable one.
{"title":"Efficient isochronous traffic management in DQDB","authors":"S. Tsai, Tein-Hsiang Lin, C. Qiao","doi":"10.1109/ICC.1995.524237","DOIUrl":"https://doi.org/10.1109/ICC.1995.524237","url":null,"abstract":"The pre-arbitrated scheme in the DQDB network is suitable for constant-rate and connection-oriented transmission. The authors propose a periodic assignment scheme, which classifies isochronous connections into several harmonic classes according to their periods. The main advantage of the proposed scheme is that by assigning the same unit channel to connections of the same class, these connections can be established as long as the total required bandwidth is no greater than the available bandwidth of the unit channel. The authors study three assignment policies based on the harmonic case of the pinwheel algorithm. These assignment policies are evaluated in terms of their blocking probabilities through both simulation and analysis. It is found that the busy-first policy, which tries to assign a new isochronous connection with a unit channel already carrying other connections of the same class, is the most suitable one.","PeriodicalId":241383,"journal":{"name":"Proceedings IEEE International Conference on Communications ICC '95","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1995-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115388114","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Coherent phase shift keying (C-PSK) represents an efficient modulation scheme for satellite communications. In such systems, carrier phase synchronizers often implement maximum likelihood (ML) estimation algorithms, the reason for this being the need of robust modem operations at low signal-to-noise ratios and the availability of digital technology. In this paper, performance of a feed-forward decision-feedback (FF-DFB) ML carrier phase synchronizer for M-PSK signals (M/spl les/18) on the AWGN channel is investigated. Open loop characteristics are analytically derived. A novel equivalent model for the FF-DFD synchronizer is presented and used to derive synchronizer tracking performance.
{"title":"Feed-forward decision-feedback carrier phase synchronizer for M-PSK signals","authors":"R. De Gaudenzi, V. Vanghi, T. Garde","doi":"10.1109/ICC.1995.524262","DOIUrl":"https://doi.org/10.1109/ICC.1995.524262","url":null,"abstract":"Coherent phase shift keying (C-PSK) represents an efficient modulation scheme for satellite communications. In such systems, carrier phase synchronizers often implement maximum likelihood (ML) estimation algorithms, the reason for this being the need of robust modem operations at low signal-to-noise ratios and the availability of digital technology. In this paper, performance of a feed-forward decision-feedback (FF-DFB) ML carrier phase synchronizer for M-PSK signals (M/spl les/18) on the AWGN channel is investigated. Open loop characteristics are analytically derived. A novel equivalent model for the FF-DFD synchronizer is presented and used to derive synchronizer tracking performance.","PeriodicalId":241383,"journal":{"name":"Proceedings IEEE International Conference on Communications ICC '95","volume":"94 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1995-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115415295","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
We describe the theory, design and implementation of a 2.5-Gb/s ATM switch growable to very large sizes, say 512/spl times/512 (1.28 Tb/s capacity), using a concentrator-based growable switch architecture. This architecture is based on a front-end concentrator-based cell distribution network followed by fixed-size ATM output switches. The front-end concentrator arrangement provides FIFO cell distribution with practical modularity and negligible cell loss. The ATM output switches are small-size shared memory designs that ensure optimal delay throughput performance for arbitrary traffic patterns. Our prototype consists of a complete and independent 2.5-Gb/s 8/spl times/8 ATM switch system with optical 2.4-Gb/s OC-48c and 155 Mb/s OC-3c interfaces, and also an expansion module (half shelf) capable of supporting eight such 8/spl times/8 switch systems yielding a total prototype capacity of 160 Gb/s.
{"title":"A 160-Gb/s ATM switch prototype using the concentrator-based growable switch architecture","authors":"K. Eng, M. Karol, G. Cyr, M.A. Pashan","doi":"10.1109/ICC.1995.525228","DOIUrl":"https://doi.org/10.1109/ICC.1995.525228","url":null,"abstract":"We describe the theory, design and implementation of a 2.5-Gb/s ATM switch growable to very large sizes, say 512/spl times/512 (1.28 Tb/s capacity), using a concentrator-based growable switch architecture. This architecture is based on a front-end concentrator-based cell distribution network followed by fixed-size ATM output switches. The front-end concentrator arrangement provides FIFO cell distribution with practical modularity and negligible cell loss. The ATM output switches are small-size shared memory designs that ensure optimal delay throughput performance for arbitrary traffic patterns. Our prototype consists of a complete and independent 2.5-Gb/s 8/spl times/8 ATM switch system with optical 2.4-Gb/s OC-48c and 155 Mb/s OC-3c interfaces, and also an expansion module (half shelf) capable of supporting eight such 8/spl times/8 switch systems yielding a total prototype capacity of 160 Gb/s.","PeriodicalId":241383,"journal":{"name":"Proceedings IEEE International Conference on Communications ICC '95","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1995-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115632673","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The paper presents a self-organizing, multihop packet radio network architecture for multimedia. This architecture relies on the code division access scheme. Nodes are organized into subsets called clusters. Within a cluster, any two nodes are one hop away from each other. All links in a cluster are assigned a common spread-spectrum code. Channel access in a cluster is implemented using a round robin scheme for data traffic and a piggyback scheme with bandwidth reservation for real time traffic. At call setup time, this architecture constructs virtual circuits and allocates bandwidth to each real time session. This multicluster architecture provides a good spatial reuse of the bandwidth. Since none of the nodes need to keep accurate common time, it eliminates the overhead of a global clock or time synchronization mechanism.
{"title":"A distributed control scheme in multi-hop packet radio networks for voice/data traffic support","authors":"C. Richard, Linand Mario Gerla","doi":"10.1109/ICC.1995.524297","DOIUrl":"https://doi.org/10.1109/ICC.1995.524297","url":null,"abstract":"The paper presents a self-organizing, multihop packet radio network architecture for multimedia. This architecture relies on the code division access scheme. Nodes are organized into subsets called clusters. Within a cluster, any two nodes are one hop away from each other. All links in a cluster are assigned a common spread-spectrum code. Channel access in a cluster is implemented using a round robin scheme for data traffic and a piggyback scheme with bandwidth reservation for real time traffic. At call setup time, this architecture constructs virtual circuits and allocates bandwidth to each real time session. This multicluster architecture provides a good spatial reuse of the bandwidth. Since none of the nodes need to keep accurate common time, it eliminates the overhead of a global clock or time synchronization mechanism.","PeriodicalId":241383,"journal":{"name":"Proceedings IEEE International Conference on Communications ICC '95","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1995-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123988892","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
A compression technique is required to increase the efficiency of the communication channel. The LZW (Lempel Ziv Welch) algorithm is used as a compression algorithm in communication channel. The LZW algorithm, however, has some redundancies in Hangeul text and voice. The VDI-LZW (voice data integrated LZW) algorithm that decreases the redundancies of the LZW algorithm is suggested as an efficient compression method of Hangeul text and voice. The VDI-LZW algorithm uses both the codeword of the complete Hangeul and a variable length codeword methodology in data compression. For voice compression, it uses the differential method that reduces the codeword size as well as the length of repeated string. This increases the repetition ratio. According to the simulation results, it can be seen that the performance of the proposed compression algorithm is better by 4% to 18% in the data compression ratio and by 35% to 44% in the voice compression ratio than that of the conventional modified LZW algorithms.
{"title":"A study on the efficient compression algorithm of the voice/data integrated multiplexer","authors":"Gyoun-Yon Cho, D. Cho","doi":"10.1109/ICC.1995.524441","DOIUrl":"https://doi.org/10.1109/ICC.1995.524441","url":null,"abstract":"A compression technique is required to increase the efficiency of the communication channel. The LZW (Lempel Ziv Welch) algorithm is used as a compression algorithm in communication channel. The LZW algorithm, however, has some redundancies in Hangeul text and voice. The VDI-LZW (voice data integrated LZW) algorithm that decreases the redundancies of the LZW algorithm is suggested as an efficient compression method of Hangeul text and voice. The VDI-LZW algorithm uses both the codeword of the complete Hangeul and a variable length codeword methodology in data compression. For voice compression, it uses the differential method that reduces the codeword size as well as the length of repeated string. This increases the repetition ratio. According to the simulation results, it can be seen that the performance of the proposed compression algorithm is better by 4% to 18% in the data compression ratio and by 35% to 44% in the voice compression ratio than that of the conventional modified LZW algorithms.","PeriodicalId":241383,"journal":{"name":"Proceedings IEEE International Conference on Communications ICC '95","volume":"56 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1995-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124107912","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Changcheng Huang, M. Devetsikiotis, I. Lambadaris, A. Kaye
Self-similar (or fractal) stochastic processes were proposed as more accurate models of certain categories of traffic (e.g., Ethernet traffic, variable-bit-rate video) which will be transported in ATM networks. Existing analytical results for the tail distribution of the waiting time in a single server queue based on fractional Gaussian noise and large deviation theory, are valid under a steady-state regime and for an asymptotically large buffer size. However, the predicted performance based on steady-state regimes may be overly pessimistic for practical applications. Theoretical approaches used to obtain the transient queueing behavior and queueing distributions for a small buffer size become quickly intractable. The approach we followed was based on fast simulation techniques for the study of certain rare events such as cell losses with very small probability of occurrence. Our simulation experiments provide an insight on the transient behavior that is not possible to predict using current analytical results. Finally they show good agreement with existing results when approaching steady-state.
{"title":"Fast simulation for self-similar traffic in ATM networks","authors":"Changcheng Huang, M. Devetsikiotis, I. Lambadaris, A. Kaye","doi":"10.1109/ICC.1995.525208","DOIUrl":"https://doi.org/10.1109/ICC.1995.525208","url":null,"abstract":"Self-similar (or fractal) stochastic processes were proposed as more accurate models of certain categories of traffic (e.g., Ethernet traffic, variable-bit-rate video) which will be transported in ATM networks. Existing analytical results for the tail distribution of the waiting time in a single server queue based on fractional Gaussian noise and large deviation theory, are valid under a steady-state regime and for an asymptotically large buffer size. However, the predicted performance based on steady-state regimes may be overly pessimistic for practical applications. Theoretical approaches used to obtain the transient queueing behavior and queueing distributions for a small buffer size become quickly intractable. The approach we followed was based on fast simulation techniques for the study of certain rare events such as cell losses with very small probability of occurrence. Our simulation experiments provide an insight on the transient behavior that is not possible to predict using current analytical results. Finally they show good agreement with existing results when approaching steady-state.","PeriodicalId":241383,"journal":{"name":"Proceedings IEEE International Conference on Communications ICC '95","volume":"144 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1995-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116413577","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper describes approaches for shaping and smoothing traffic from VBR MPEG video encoders and interfacing to ATM networks, by introducing delay and rate buffering between the video encoding and decoding processes, and by flow controlling the data rate out of the encoder buffer into the network. Three algorithms for flow control are defined and compared in terms of their resulting traffic profile. The performance of these algorithms in the presence of a leaky bucket ATM network access controller is studied. The relationship between the leaky bucket algorithm parameters and VBR MPEG encoder parameters is discussed. It is shown that significant gains in performance can be obtained by appropriate selection of traffic smoothing algorithms at the output of a VBR MPEG encoder.
{"title":"Source traffic smoothing and ATM network interfaces for VBR MPEG video encoders","authors":"K. Joseph, D. Reininger","doi":"10.1109/ICC.1995.524502","DOIUrl":"https://doi.org/10.1109/ICC.1995.524502","url":null,"abstract":"This paper describes approaches for shaping and smoothing traffic from VBR MPEG video encoders and interfacing to ATM networks, by introducing delay and rate buffering between the video encoding and decoding processes, and by flow controlling the data rate out of the encoder buffer into the network. Three algorithms for flow control are defined and compared in terms of their resulting traffic profile. The performance of these algorithms in the presence of a leaky bucket ATM network access controller is studied. The relationship between the leaky bucket algorithm parameters and VBR MPEG encoder parameters is discussed. It is shown that significant gains in performance can be obtained by appropriate selection of traffic smoothing algorithms at the output of a VBR MPEG encoder.","PeriodicalId":241383,"journal":{"name":"Proceedings IEEE International Conference on Communications ICC '95","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1995-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123618601","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Consider two sets of points X, A and their n-fold cartesian products {A}/sup n/, {X}n. A non-negative cost is associated with each element of A. A measure of distance is defined between an element of A and an element of X. It is assumed that the cost and also the distance in the n-fold space has an additive property. The shaped set is composed of a subset of elements of {A}/sup n/ of the least cost. Decoding of an element x/spl isin/{X}/sup n/ is the process of finding the element of the shaped set which has the minimum distance to x. Using the additivity property of cost and distance measures, the decoding problem is formulated as a linear program. Using the generalized upper bounding technique of linear programming in conjunction with some special features of the problem, we present methods to substantially reduce the complexity of the corresponding simplex search. The proposed method is used for the fixed-rate entropy-coded vector quantization of a Gaussian source. For n=128 (space dimensionality) using 8 points per dimension and for a rate of 2.5 bits/dimension, we need about 52 additions, 87 comparisons, 0.2 divisions, and 0.4 multiplications per dimension to achieve SNR=13.31 dB (the bound obtained from the rate-distortion curve is 13.52 dB). This is substantially less complex than the traditional methods based on the dynamic programming.
{"title":"Fixed-rate entropy-coded vector quantization using linear (zero-one) programming","authors":"A. Khandani","doi":"10.1109/ICC.1995.524439","DOIUrl":"https://doi.org/10.1109/ICC.1995.524439","url":null,"abstract":"Consider two sets of points X, A and their n-fold cartesian products {A}/sup n/, {X}n. A non-negative cost is associated with each element of A. A measure of distance is defined between an element of A and an element of X. It is assumed that the cost and also the distance in the n-fold space has an additive property. The shaped set is composed of a subset of elements of {A}/sup n/ of the least cost. Decoding of an element x/spl isin/{X}/sup n/ is the process of finding the element of the shaped set which has the minimum distance to x. Using the additivity property of cost and distance measures, the decoding problem is formulated as a linear program. Using the generalized upper bounding technique of linear programming in conjunction with some special features of the problem, we present methods to substantially reduce the complexity of the corresponding simplex search. The proposed method is used for the fixed-rate entropy-coded vector quantization of a Gaussian source. For n=128 (space dimensionality) using 8 points per dimension and for a rate of 2.5 bits/dimension, we need about 52 additions, 87 comparisons, 0.2 divisions, and 0.4 multiplications per dimension to achieve SNR=13.31 dB (the bound obtained from the rate-distortion curve is 13.52 dB). This is substantially less complex than the traditional methods based on the dynamic programming.","PeriodicalId":241383,"journal":{"name":"Proceedings IEEE International Conference on Communications ICC '95","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1995-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117079334","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：