Search

Given a start point s, a target point t, and k disjoint convex polygons in the given order in the plane, finding the shortest path of visiting the convex polygons sequence from s to t is the goal. In this paper, we present a fast algorithm to compute the shortest path based on the last step shortest path maps. Firstly, we locate the dividing-points quickly in the two adjacent polygons, which can reduce amounts of iteratively computation. Then, we convert this problem to solving the shortest path of visiting the disjoint line segments sequence which makes the solution much simpler. Furthermore, we analyze the complexity of new algorithm and obtain that it is superior to the previous solution. Finally, we implement this algorithm and show that it is correct.

The B. Steiner estimator used in the TD-SCDMA system assumes that the channel response during one timeslot is constant, which results in big errors under the fast varying channel condition compared with the real channel response. In order to solve this problem, a novel channel estimation method is proposed for the TD-SCDMA system. Every timeslot is separated into several segments, the multi-segment estimation and multi-slot LSM fit is used to improve the accuracy of channel estimation. By comparing this two methods and analyzing their effects to detection, especially for the joint detection (JD) algorithm of TD-SCDMA system, simulation results show that the novel channel estimation is more close to the real channel response, which can largely improve the system performance, especially for the fast varying channel environment

According to the features of TD-SCDMA system, a novel multi-antenna pre-processing model and algorithm is proposed. Meanwhile, the more precise uplink channel estimation and downlink channel prediction are put forward for TD-SCDMA system to improve the performance of basic multi-antenna pre-processing algorithm. Simulation results show that: by using the proposed multi-antenna pre-processing technology, the capacity and performance of TD-SCDMA system is significantly improved. Meanwhile, the improved multi-antenna pre-processing algorithm performs better than the basic algorithm under fast varying channel condition.

An adaptive hybrid MIMO scheme has been proposed for multi-antenna system based on the uniformed hybrid MIMO model combining the transmit diversity and spatial multiplexing. The adaptive hybrid MIMO scheme adaptively adjusts the allocation of freedom degree i.e. gain ratio between the diversity and multiplexing with the adaptive hybrid MIMO precoding according to the variation of channel condition to achieve the necessary diversity gain and spatial multiplexing gain simultaneously; Meanwhile, an adaptive hybrid detection method is designed to complete the hybrid spatial multiplexing signal separation and transmit diversity signal decoding. Simulation results show that: for the multi-antenna system with the fixed antennas, compared with other transmit diversity and spatial multiplexing schemes, the proposed adaptive hybrid MIMO scheme can achieve the best performance and highest capacity during the whole channel SNR variation span.

A downlink grouping transmission scheme has been proposed for TDD OFDM system. The coded bits are separated into several branches according to their priority, and then the results are adaptively allocated and transmitted on the 2-demension time-frequency channel blocks, which are also separated into different groups according to their power. Simulation results show that this scheme can significantly improve the performance of TDD OFDM system. Meanwhile, the grouping modulation and downlink time-frequency channel prediction can further enhance the grouping transmission.

Based on the burst structure of TD-SCDMA system, a novel downlink SDM-JT model is put forward using the channel correlation between the uplink and downlink in TDD system. Meanwhile, by analyzing the features of wireless fading channel, the basic and improved SDM-JT algorithms are proposed. Simulation results show that the SDM-JT scheme can significantly improve the capacity and performance of TD-SCDMA system while the complexity of Mobile station (MS) is sharply reduced.