-
RSSI Fingerprinting-based Localization Using Machine Learning in LoRa Networks
Authors:
Mahnoor Anjum,
Muhammad Abdullah Khan,
Syed Ali Hassan,
Aamir Mahmood,
Hassaan Khaliq Qureshi,
Mikael Gidlund
Abstract:
The scale of wireless technologies penetration in our daily lives, primarily triggered by the Internet-of-things (IoT)-based smart cities, is beaconing the possibilities of novel localization and tracking techniques. Recently, low-power wide-area network (LPWAN) technologies have emerged as a solution to offer scalable wireless connectivity for smart city applications. LoRa is one such technology…
▽ More
The scale of wireless technologies penetration in our daily lives, primarily triggered by the Internet-of-things (IoT)-based smart cities, is beaconing the possibilities of novel localization and tracking techniques. Recently, low-power wide-area network (LPWAN) technologies have emerged as a solution to offer scalable wireless connectivity for smart city applications. LoRa is one such technology that provides energy efficiency and wide-area coverage. This article explores the use of intelligent machine learning techniques, such as support vector machines, spline models, decision trees, and ensemble learning, for received signal strength indicator (RSSI)-based ranging in LoRa networks, on a training dataset collected in two different environments: indoors and outdoors. The suitable ranging model is then used to experimentally evaluate the accuracy of localization and tracking using trilateration in the studied environments. Later, we present the accuracy of LoRa-based positioning system (LPS) and compare it with the existing ZigBee, WiFi, and Bluetooth-based solutions. In the end, we discuss the challenges of satellite-independent tracking systems and propose future directions to improve accuracy and provide deployment feasibility.
△ Less
Submitted 1 June, 2020;
originally announced June 2020.
-
Adaptive System Identification using Markov Chain Monte Carlo
Authors:
Muhammad Ali Raza Anjum
Abstract:
One of the major problems in adaptive filtering is the problem of system identification. It has been studied extensively due to its immense practical importance in a variety of fields. The underlying goal is to identify the impulse response of an unknown system. This is accomplished by placing a known system in parallel and feeding both systems with the same input. Due to initial disparity in thei…
▽ More
One of the major problems in adaptive filtering is the problem of system identification. It has been studied extensively due to its immense practical importance in a variety of fields. The underlying goal is to identify the impulse response of an unknown system. This is accomplished by placing a known system in parallel and feeding both systems with the same input. Due to initial disparity in their impulse responses, an error is generated between their outputs. This error is set to tune the impulse response of known system in a way that every change in impulse response reduces the magnitude of prospective error. This process is repeated until the error becomes negligible and the responses of both systems match. To specifically minimize the error, numerous adaptive algorithms are available. They are noteworthy either for their low computational complexity or high convergence speed. Recently, a method, known as Markov Chain Monte Carlo (MCMC), has gained much attention due to its remarkably low computational complexity. But despite this colossal advantage, properties of MCMC method have not been investigated for adaptive system identification problem. This article bridges this gap by providing a complete treatment of MCMC method in the aforementioned context.
△ Less
Submitted 8 May, 2015;
originally announced May 2015.
-
A New Approach to Linear Estimation Problem in Multi-user Massive MIMO Systems
Authors:
Muhammad Ali Raza Anjum
Abstract:
A novel approach for solving linear estimation problem in multi-user massive MIMO systems is proposed. In this approach, the difficulty of matrix inversion is attributed to the incomplete definition of the dot product. The general definition of dot product implies that the columns of channel matrix are always orthogonal whereas, in practice, they may be not. If the latter information can be incorp…
▽ More
A novel approach for solving linear estimation problem in multi-user massive MIMO systems is proposed. In this approach, the difficulty of matrix inversion is attributed to the incomplete definition of the dot product. The general definition of dot product implies that the columns of channel matrix are always orthogonal whereas, in practice, they may be not. If the latter information can be incorporated into dot product, then the unknowns can be directly computed from projections without inverting the channel matrix. By doing so, the proposed method is able to achieve an exact solution with a 25% reduction in computational complexity as compared to the QR method. Proposed method is stable, offers an extra flexibility of computing any single unknown, and can be implemented in just twelve lines of code.
△ Less
Submitted 28 April, 2015;
originally announced April 2015.
-
A New Approach to Adaptive Signal Processing
Authors:
Muhammad Ali Raza Anjum
Abstract:
A unified linear algebraic approach to adaptive signal processing (ASP) is presented. Starting from just Ax=b, key ASP algorithms are derived in a simple, systematic, and integrated manner without requiring any background knowledge to the field. Algorithms covered are Steepest Descent, LMS, Normalized LMS, Kaczmarz, Affine Projection, RLS, Kalman filter, and MMSE/Least Square Wiener filters. By fo…
▽ More
A unified linear algebraic approach to adaptive signal processing (ASP) is presented. Starting from just Ax=b, key ASP algorithms are derived in a simple, systematic, and integrated manner without requiring any background knowledge to the field. Algorithms covered are Steepest Descent, LMS, Normalized LMS, Kaczmarz, Affine Projection, RLS, Kalman filter, and MMSE/Least Square Wiener filters. By following this approach, readers will discover a synthesis; they will learn that one and only one equation is involved in all these algorithms. They will also learn that this one equation forms the basis of more advanced algorithms like reduced rank adaptive filters, extended Kalman filter, particle filters, multigrid methods, preconditioning methods, Krylov subspace methods and conjugate gradients. This will enable them to enter many sophisticated realms of modern research and development. Eventually, this one equation will not only become their passport to ASP but also to many highly specialized areas of computational science and engineering.
△ Less
Submitted 23 April, 2015;
originally announced April 2015.
