摘要：An incremental redundancy hybrid automatic repeat- request （IR-HARQ）scheme based on irregular repeat-accumulate （IRA）codes is proposed. The design of rate compatible punctured IRA codes suitable for an IR-HARQ scheme is well formulated and efficiently solved by a linear-programming method, along with a one-dimensional approach for density evolution. Compared to IR-HARQ schemes based on turbo codes, simulation shows that the proposed IR-HARQ schemes based on IRA codes may achieve almost the same performance at a block size of 1 024, but better throughput at a block size of 4 096. The advantages of the proposed scheme in implementation, including decoding complexity and parallelism, make it more attractive in practice than the IR-HARQ schemes based on both turbo and LDPC codes.

摘要：In this paper a method that combines transmit antenna selection and reduced-constellation detection in spatially correlated Multi-Input Multi-Output （MIMO） fading channels is presented. To mitigate the performance degradation caused by the use of antenna selection that is based on correlation among columns, an iterative receiver scheme that uses only a subset of the constellation points close to the expected symbol vahle estimated in the previous iteration is proposed. The size of the subset can adapt to the maximum correlation of the sub-matrix after the simple antenna selection. Furthermore, the error rate performance of the scheme under linear Miniinutn Mean Square Error （MMSE） or Ordered Successive Interference Cancellation （OSIC） for the first run detection and different interleaver lengths is investigated while the transnlit antenna selection is considered. The simulation results show a significant advantage both for implementation complexity and for error rate performance under a fixed data rate.

摘要：广义多载波并行传输方案是一种宽子带方案,每个子带带宽达到1.28MHz且子带交叠在-15dB以下.这种多载波结构能够有效地克服多载波系统的峰均比问题和对频偏的敏感度.首先给出了广义多载波并行传输方案的数字基带模型.为了便于信道估计和信号检测,设计了一种双循环的自适应时隙结构,基于该结构设计了频域均衡空时联合检测器,并证明它是最小均方误差准则下的最优检测器.最后在ITU M.1225信道下计算机仿真证明了该方案在宽带移动通信环境中的有效性.

摘要：提出一款可以工作在极低电源电压条件下,功耗极低的亚阈值SRAM存储单元.为使本设计在极低电源电压(200 mV)条件下依然能够保持足够的鲁棒性,采用差分读出方式和可配置的操作模式.为极大限度地降低电路功耗,采用自适应泄漏电流切断机制,该机制在不提高动态功耗与不增加性能损失的前提下,可同时降低动态操作(读/写操作)和静态操作时的泄漏电流.基于IBM 130 nm工艺,实现了一款256×32 bit大小的存储阵列.测试结果表明,该存储阵列可以在200 mV电源电压条件下正常工作,功耗(包括动态功耗和静态功耗)仅0.13μW,为常规六管存储单元功耗的1.16%.

摘要：With more scaling, the speed of than 40 years Moore CMOS transistors is around 100 GHz. Such fact makes it possible to realize mm-wave circuits in CMOS. However, with the target of achieving broadband and power-efficient operation, 60 GHz CMOS RF transceiver faces severe challenges. After reviewing the technology issues, regarding the 60 GHz applications, this paper discusses design challenges both from the system and the building block levels, and also presents some simulated or measured circuits results.

摘要：本文根据ＩＳ－９５上行链路标准，分析了基站ＲＡＫＥ接收机设计原理，讨论了其实现方法，完成了整个电路的ＶＨＤＬ设计、验证以及ＦＰＧＡ实现，并给出了主要的设计结果。本文的工作对于我国ＣＤＭＡ移动通信系统的设计和研制具有重要意义。

摘要：Based on the frequency domain training sequences, the polynomial-based carrier frequency offset （CFO） estimation in multiple-input multiple-output （ MIMO ） orthogonal frequency division multiplexing （ OFDM ） systems is extensively investigated. By designing the training sequences to meet certain conditions and exploiting the Hermitian and real symmetric properties of the corresponding matrices, it is found that the roots of the polynomials corresponding to the cost functions are pairwise and that both meger CFO and fractional CFO can be estimated by the direct polynomial rooting approach. By analyzing the polynomials corresponding to the cost functions and their derivatives, it is shown that they have a common polynomial factor and the former can be expressed in a quadratic form of the common polynomial factor. Analytical results further reveal that the derivative polynomial rooting approach is equivalent to the direct one in estimation at the same signal-to-noise ratio（SNR） value and that the latter is superior to the former in complexity. Simulation results agree well with analytical results.

摘要：设计了一种采用预编码方式的准正交空时编码（quasi-orthogonal space time code, QOSTC）,其编码方案是首先对发送信号向量进行与信道状态信息无关的酉变换,然后再进行正交编码.酉变换可将QOSTC信道能转化为并行复用传输信道,实现无符号间干扰的高速率并行传输,且接双端信号的最大似然（maximum likelihood,ML）符号译码可简化成线性译码,从而可采用简单的线性检测降低译码复杂度.仿真结果表明,在高信噪比下,酉变换的QOSTC线性检测比没有酉变换的QOSTC的ML检测差2dB左右,但其分集度几乎相同.

摘要：研究了长期演进(long term evolution,LTE)网络中考虑机会调度并可提供不同服务质量(QoS)保障的准入控制(CAC)算法的设计.使用基于累积分布函数的调度(cumulative distributed function based scheduling,CS)作为基本调度策略.首先采用机会轮询(opportunistic roundrobin,ORR)计算CS性能下界,并通过仿真验证此计算方法的正确性.然后提出一个CS与ORR相结合的CAC算法COCDQ(CS/ORR based CAC algorithm for different QoS requirements),其可同时保障新接入用户和已存在用户的不同QoS需求.最后通过系统级仿真验证所提算法性能,结果表明联合考虑机会调度与QoS需求,COCDQ算法可有效降低新接入用户阻塞率,提供更好的QoS保障,其代价是仅总吞吐量略有降低.

摘要：Based on the transform-domain characteristics of pilot signals,a band suppression filter is used as a transform-domain filter to restrain the interference of noise in channel *** performance effect on channel estimation for an orthogonal frequency division multiplex （OFDM） system by different energy coefficients in the transform domain and the energy coefficient under the different signal-to-noise ratios （SNR） are also analyzed.A new energy coefficient expression is *** is theoretically proven that dynamically selecting an energy coefficient can significantly improve the performance of channel *** results show that the proposed algorithm can achieve better performance close to the theoretic bounds of perfect channel estimation. The algorithm is adapted to single-input single-output （SISO） OFDM and multi-input multi-output （MIMO） OFDM systems.