David J. Y. Lee

Tunneling wireless voice with software defined vocoder

The objective of these new algorithms is to specify the means of reducing the vocoding conversions among different PLMN, PSTN and IP networks. This invention is one of the key enabler to the wireless VoIP application. VoIP is getting more and more momentum and is viewed as the future way of transporting voice. This new algorithm uses the integrated signaling among wireline and wireless (PSTN/PLMN, TDMA, GSM, CDMA, AMPS) networks so that wireless voice can be tunneled through different wireless system without any vocoding (except at the handset). This provides a means of efficient bandwidth, vocoding resources and potential delay management for wireless VoIP application. After the calling party dialed the digits, the signaling will be set up to notify the called party network that the call is on which wireless technology and how the network and handset will be handling the call. The called party network can have local storage media to load the vocoder driver if the software defined handset concept is adopted. As 3G is right around the corner, some features that could be supported by 3G are also discussed. For example, more efficient and intelligent interaction between the user and handset can be a reality now. The fundamental channel can be used to control the air-link resources and become more flexible and efficient.

Enhanced Lee Model from Rough Terrain Sampling Data Aspect

The Lee propagation prediction model has been well recognized by the wireless industry as one of the most accurate propagation prediction models. This paper discusses innovative approaches dealing with rough digital samples of terrain data and enhancements to the Lee model during the validation process. In general, the Lee model is composed of two parts, the impact of man-made structures and the impact of the natural terrain variation [1-6]. There are other papers discuss innovative algorithms on calculating effective antenna gain [7,8] and diffraction loss [7-9] as well as on enhancing the Lee model [12-14]. This paper focuses on the natural (terrain) factor. The new algorithm presented in this paper is quite different than others as it integrates both Line Of Site (LOS) and shadow loss calculation together for the Lee model. First, in the LOS scenario, it addresses the issue of big swing of effective antenna gain due to non-continuous terrain data. Second, in the non-LOS situation, the effective antenna gain is integrated with shadow loss. Both single knife edge and multiple knife edge scenarios are discussed. The new algorithm is developed based on the analysis of measured and predicted (the theoretical shadow loss and effective antenna gain) data. The new algorithm involves more calculation but it improves the accuracy of the predicted value. This algorithm was implemented and verified using field terrain and measurement data from a variety of different environments and different countries including Italy, US, Spain, Japan, South Korea, Taiwan and Romania.

IP voice mail

The objective of these new algorithms is to specify means of using IP to leave and retrieve voice mail. The next generation wireless/wireline network is targeted to use IP as the transport for data and voice. However, the integration of PLMN/PSTN voice mail with IP network is not fully specified. Traditionally, the voice packet always has priority in the transmission in the IP network. The voice mail can and should be handled in a more efficient means so that its quality and timing will not suffer. However, the cost can be significantly reduced. This proposal uses the integrated signaling among wireline and wireless (PSTN/PLMN, TDMA, GSM, CDMA, AMPS) networks so that voice mail can be stored and later forward to the final destination with appropriate and most efficient bandwidth and vocoding resources. After the called party voicemail engine takes over, the signaling will be set up to notify the calling party network that the call is going to voice mail system. The calling party network can have local storage media to take over the recording from the caller and gives a estimated time of delivery on the current network traffic. The voice packets will, then, be sent through the IP network. It works like store and forward fax with some kind of efficiency and intelligence built in. IP multicast technology can also be integrated to support group or broadcast voice mail. As 3G is right around the corner, some features that could be supported by 3G technology are also discussed. For example, more efficient and intelligent interaction between user and handset regarding voice mail can be a reality now. The fundamental channel can be used to control the air-link resources. The JAVA enabled handset can support voice mail applications and be the middleware between user and system. This provides a means to better use the RF resources.

Enhanced Lee In-Building Model

This paper provides a complete solution for building related propagation, starting by validating the Lee In-building model [1-3] for the single floor scenario. Then how the enhanced Lee In-building model performs in inter-floor and inter-building cases are discussed. Measurement data were collected from 2.4G WLAN developed systems installed in many different floors in two different buildings which are connected with a corridor. The data are from multiple single floor measurements, inter floor measurement and inter building measurements. The FDTD and Ray Tracing models [4-9] was used to validate the measured with the Lee model. The results show that the Lee model outperforms the FDTD and Ray Tracing models in both speed and accuracy. The propagation environment is quite unique and challenging for this deployment. The Lee model was used to optimize the coverage and minimize the interference while fine tuning radio parameters (antenna types, ERP, down tilt) to improve system performances. The net results were continuous connectivity with smooth roaming and high throughput which provided a different experience for users.

Resource reuse for 3G WOS

The wireless office system (WOS) is gaining momentum in the wireless industry; how to use 3G resources effectively to deploy an optimized WOS system is the focus of our proposal. We first consider the concept and implementation of code reuse in 3G WOS. To compete with wireline and to ensure the success of wireline migration to wireless, the MOU needs to be substantially increased for both home and enterprise usage (in-building). The code reuse concept captures the idea of reusing the code for the enterprise network. This can resolve the intra cell interference issue. We then consider the timing offset reuse concept for resolving the inter cell interference issues. By combining these two factors, the capacity and available resource for enterprise can be increased multifold. We also discuss how a better location identification algorithm can be achieved through sharing these resources. The unique way of using cell breathing to identify a user is also discussed. A smaller timing offset is introduced for enterprise PN. This enables efficient deployment of in-building picocells. Inventions included in this proposal are CDMA radio resource reuse (include code reuse and time offset reuse) and an associated location algorithm for better performance and location identification in-building.

Integrated Models and Their Usage in Predicting the Signal Strength

There are many propagation models that address specific area and applications needs. Each model works well in specific application with specific data and algorithm. Mobile users are now expecting more and more ubiquitous user experience. It is critical to have an integrated model that provides a complete solution for deployment and optimization for different environment and cell types. The Lee macro, micro and pico propagation models which covers different frequencies (900MHz, 1800 MHz, 2.4G and 5 G) and are validated with large amount of field data have been adopted in many deployments [1-7]. In this paper we integrate these three models to be used for deploying, and optimizing system coverage, interference, capacity and handoff across different cell types. Systems are designed and deployed with common input, algorithm and output. It is integrated with common measured data also to enhance the accuracy and flexibility of the model with reduced cost. Each Lee sub model is reviewed and compared with measured data to prove its accuracy. Then the integrated Lee model is used to do a complete system design with pico, micro and macro cells. The design is then compared with the ray tracing and Finite-difference time-domain (FDTD) techniques [8-10] to validate its applicability. Results show that integrated Lee models do provide an effective and accurate way for a complete system design.

4D morphology model with feedback loop

This paper addresses innovative ways of dealing with the impact of morphology on network engineering. There are many different morphologies and each has unique impact on the network engineering and planning. For example, different types of trees will have different impact on the propagation. Also, it behaves differently if it is weather sensitive. In this invention, a 4D (time of the year, height, width and length of the morphology) database and associated algorithm are proposed to deal with the impact on system engineering. The key point of this algorithm is that it includes means to feedback and integrate measured data from the field so that the 4D morphology model can be tuned (supports full flexibility) to better handle different situations. Inventions include first the fundamental concept of dealing with morphology in a integrated way with system propagation. The morphology specific slope will be derived and applied to the existing model. Second, is the algorithm of deriving these morphology slopes. Third, is to include the morphology in the calculation of shadow loss. Fourth, is to identify and integrate the morphology impact on time variance and associated engineering algorithm.

Integrating mobile with IP phone

The objective of this proposal is to specify the algorithm to integrate mobile and LAN/WLAN with IP phone so that wireless system can work seamlessly together with a wired system. By implementing this proposal, the mobile phone can be power charged by IP, more granular location identification by IP, seamless roam and HO among wireless and wired IP networks. Calls can be routed based on the location, most convenient and/or cost effective way of connection.