The design of a framework to control multiple semi-autonomous mobile robots is presented. The goals are to create a flexible framework that allows easy task reconfiguration for a fleet of mobile robots, and to provide a means of off-loading computationally intensive tasks to a central server. Four major components are specified: controllable devices, application servers, an application directory, and a mobile management system. Remote firmware upgrades via the mobile management system are achieved with the remotely managed operating system framework or RMOS. A prototype was created using field programmable gate arrays and system on a programmable chip technology from Altera. It was built with the application of selective herbicide spraying in mind. Although the prototype was a successful implementation of the designed framework, selective herbicide application was not implemented. A broad-leaf weed detection algorithm using Boltzmann machines was investigated. Its performance was good, but similar to that obtained by a statistical analysis of the input images of a lawn.

While Application Specific Instruction Set Processors (ASIPs) have allowed designers to create processors with custom instructions to target specific applications, floating-point units (FPUs) are still instantiated as non-customizable general-purpose units, which if under utilized, wastes area and performance. However, customizing FPUs manually is a complex and time-consuming process. Therefore, there is a need for an automated custom FPU generation scheme. This thesis presents a methodology for generating application-specific FPUs customized at the instruction level, with integrated datapath merging to minimize area. The methodology reduces the subset of floating-point instructions implemented to the minimum required for the application. Datapath merging is then performed on the required datapaths to minimize area. Previous datapath merging techniques failed to consider merging components of different bit-widths and thus ignore the bit-alignment problem in datapath merging. This thesis presents a novel bit-alignment solution during datapath merging. In creating the custom FPU, the subset of floating-point instructions that should be implemented in hardware has to be determined. Implementing more instructions in hardware reduces the cycle count of the application, but may lead to increased delay due to multiplexers inserted on the critical path during datapath merging. A rapid design space exploration was performed to explore the trade-offs. By performing this exploration, a designer could determine the number of instructions that should be implemented as a custom FPU and the number that should be left for software emulation, such that performance and area meets the designer's requirements. Customized FPUs were generated for different Mediabench applications and compared to a fully-featured reference FPU that implemented all floating-point operations. Reducing the floating-point instruction set reduced the FPU area by an average of 55%. Performing instruction reduction and then datapath merging reduced the FPU area by an average of 68%. Experiments showed that datapath merging without bit-alignment achieved an average area reduction of 10.1%. With bit-alignment, an average of 16.5% was achieved. Bit-alignment proved most beneficial when there was a diverse mix of different bit-widths in the datapaths. Performance of Field-Programmable Gate Arrays (FPGAs) used for floating-point applications is poor due to the complexity of floating-point arithmetic. Implementing floating-point units on FPGAs consume a large amount of resources. Therefore, there is a need for embedded FPUs in FPGAs. However, if unutilized, they waste area on the FPGA die. To overcome this issue, a novel flexible multi-mode embedded FPU for FPGAs is presented in this thesis that can be configured to perform a wide range of operations. The floating-point adder and multiplier in the embedded FPU can each be configured to perform one double-precision operation or two single-precision operations in parallel. To increase flexibility further, access to the large integer multiplier, adder and shifters in the FPU is provided. It is also capable of floating-point and integer multiply-add operations. Benchmark circuits were implemented on both a standard Xilinx Virtex-II FPGA and on the FPGA with embedded FPU blocks. The implementations on the FPGA with embedded FPUs showed mean area and delay improvements of 5.2x and 5.8x respectively for the double-precision benchmarks, and 4.4x and 4.2x for the single-precision benchmarks.

While Application Specific Instruction Set Processors (ASIPs) have allowed designers to create processors with custom instructions to target specific applications, floating-point units (FPUs) are still instantiated as non-customizable general-purpose units, which if under utilized, wastes area and performance. However, customizing FPUs manually is a complex and time-consuming process. Therefore, there is a need for an automated custom FPU generation scheme. This thesis presents a methodology for generating application-specific FPUs customized at the instruction level, with integrated datapath merging to minimize area. The methodology reduces the subset of floating-point instructions implemented to the minimum required for the application. Datapath merging is then performed on the required datapaths to minimize area. Previous datapath merging techniques failed to consider merging components of different bit-widths and thus ignore the bit-alignment problem in datapath merging. This thesis presents a novel bit-alignment solution during datapath merging. In creating the custom FPU, the subset of floating-point instructions that should be implemented in hardware has to be determined. Implementing more instructions in hardware reduces the cycle count of the application, but may lead to increased delay due to multiplexers inserted on the critical path during datapath merging. A rapid design space exploration was performed to explore the trade-offs. By performing this exploration, a designer could determine the number of instructions that should be implemented as a custom FPU and the number that should be left for software emulation, such that performance and area meets the designer's requirements. Customized FPUs were generated for different Mediabench applications and compared to a fully-featured reference FPU that implemented all floating-point operations. Reducing the floating-point instruction set reduced the FPU area by an average of 55%. Performing instruction reduction and then datapath merging reduced the FPU area by an average of 68%. Experiments showed that datapath merging without bit-alignment achieved an average area reduction of 10.1%. With bit-alignment, an average of 16.5% was achieved. Bit-alignment proved most beneficial when there was a diverse mix of different bit-widths in the datapaths. Performance of Field-Programmable Gate Arrays (FPGAs) used for floating-point applications is poor due to the complexity of floating-point arithmetic. Implementing floating-point units on FPGAs consume a large amount of resources. Therefore, there is a need for embedded FPUs in FPGAs. However, if unutilized, they waste area on the FPGA die. To overcome this issue, a novel flexible multi-mode embedded FPU for FPGAs is presented in this thesis that can be configured to perform a wide range of operations. The floating-point adder and multiplier in the embedded FPU can each be configured to perform one double-precision operation or two single-precision operations in parallel. To increase flexibility further, access to the large integer multiplier, adder and shifters in the FPU is provided. It is also capable of floating-point and integer multiply-add operations. Benchmark circuits were implemented on both a standard Xilinx Virtex-II FPGA and on the FPGA with embedded FPU blocks. The implementations on the FPGA with embedded FPUs showed mean area and delay improvements of 5.2x and 5.8x respectively for the double-precision benchmarks, and 4.4x and 4.2x for the single-precision benchmarks.

abstract: With the rise of the Internet of Things, embedded systems have become an integral part of life and can be found almost anywhere. Their prevalence and increased interconnectivity has made them a prime target for malicious attacks. Today, the vast majority of embedded devices are powered by ARM processors. To protect their processors from attacks, ARM introduced a hardware security extension known as TrustZone. It provides an isolated execution environment within the embedded device in which to deploy various memory integrity and malware detection tools. Even though Secure World can monitor the Normal World, attackers can attempt to bypass the security measures to retain control of a compromised system. CacheKit is a new type of rootkit that exploits such a vulnerability in the ARM architecture to hide in Normal World cache from memory introspection tools running in Secure World by exploiting cache locking mechanisms. If left unchecked, ARM processors that provide hardware assisted cache locking for performance and time-critical applications in real-time and embedded systems would be completely vulnerable to this undetectable and untraceable attack. Therefore, a new approach is needed to ensure the correct use of such mechanisms and prevent malicious code from being hidden in the cache. CacheLight is a lightweight approach that leverages the TrustZone and Virtualization extensions of the ARM architecture to allow the system to continue to securely provide these hardware facilities to users while preventing attackers from exploiting them. CacheLight restricts the ability to lock the cache to the Secure World of the processor such that the Normal World can still request certain memory to be locked into the cache by the secure operating system (OS) through a Secure Monitor Call (SMC). This grants the secure OS the power to verify and validate the information that will be locked in the requested cache way thereby ensuring that any data that remains in the cache will not be inconsistent with what exists in main memory for inspection. Malicious attempts to hide data can be prevented and recovered for analysis while legitimate requests can still generate valid entries in the cache. / Dissertation/Thesis / Masters Thesis Computer Science 2018

abstract: Majority of the Sensor networks consist of low-cost autonomously powered devices, and are used to collect data in physical world. Today's sensor network deployments are mostly application specific & owned by a particular entity. Because of this application specific nature & the ownership boundaries, this modus operandi hinders large scale sensing & overall network operational capacity. The main goal of this research work is to create a mechanism to dynamically form personal area networks based on mote class devices spanning ownership boundaries. When coupled with an overlay based control system, this architecture can be conveniently used by a remote client to dynamically create sensor networks (personal area network based) even when the client does not own a network. The nodes here are "borrowed" from existing host networks & the application related to the newly formed network will co-exist with the native applications thanks to concurrency. The result allows users to embed a single collection tree onto spatially distant networks as if they were within communication range. This implementation consists of core operating system & various other external components that support injection maintenance & dissolution sensor network applications at client's request. A large object data dissemination protocol was designed for reliable application injection. The ability of this system to remotely reconfigure a network is useful given the high failure rate of real-world sensor network deployments. Collaborative sensing, various physical phenomenon monitoring also be considered as applications of this architecture. / Dissertation/Thesis / M.S. Computer Science 2013

The goal of the master thesis is to investigate the feasibility ofhaving distributed machine learning on embedded devices and toanalyse how the architecture of such a system can look like. A systemis proposed which enables machine learning running on multipleembedded devices to communicate with an end application. Theapplication communicates with the distributed machine learning via agateway, which decouples the application. The proposed system isimplemented as a proof of concept system, which utilizes distributedmachine learning to achieve gesture recognition. The Intel Curiemodule was selected as the embedded device, as it provides a hardwareaccelerated implementation of the machine learning algorithmsK-Nearest Neighbour and Radial Basis Function. This module alsoprovides accelerometer/gyroscope sensors for collecting gesture dataas well as Bluetooth Low Energy, which enables wireless communicationwith other devices. The implemented system shows that it is feasibleto implement distributed machine learning on embedded devices if themachine learning is hardware accelerated. If a hardware acceleratorwas not used the computational load on the embedded device willincrease, which will increase the overall power consumption. For alow powered and constrained devices, hardware accelerated machinelearning is likely the best approach to implement edge intelligence.

Unique customizations (options) of features in forklifts are often requested by customers. When new options are created or existing options have to be modified in the main software the complexity increases, the firmware revision pool gets large and with the increasing code size the memory limit is threatened. This affects the software development since the frequent modification of the option handler software is very resource consuming. Therefore it is desirable to have a highly modular system for the option handler to simplify the development process. Although the market value of this improvement is negligible the possible long term savings is the desirable effect. This thesis explores the possibility of migrating the option handling software to a dedicated hardware module. This helps the development process by increasing the modularity of the system architecture and thus reducing the development scope. The tools and the approach to accomplish this option handler is analyzed. A system model of the resulting approach is designed and a prototype is developed to validate the result.

Ericsson has always strived for the technology leadership in its offering by designing products based on the latest technology. Going ahead with a similar thought it started exploring an idea of running a mobile core application using a Simplified Infrastructure (SI) to eventually enable the Cloud based solutions. But in order to run these type of applications in the Cloud, the in-service performance provided by such a SI should be the same as the native infrastructure in order to maintain the mobile core application’s QoS. "High availability" of the infrastructure is one of the measure of the ISP and from the ISP point of view, such a migration would be considered feasible only if the SI is able to maintain the same level of availability as provided by the native infrastructure solution without bringing in any major architecture changes within the SI. Hence this master thesis project investigates the feasibility of achieving the same availability as before if the mobile core application is to be migrated from the native infrastructure to the SI. Such a feasibility exploration was the very first attempt with respect to the SI within Ericsson, which was executed through this master thesis project. In order to achieve the goal of this thesis project a detailed system study was carried out, which focused on the native infrastructure architecture, how it was maintaining the "high availability" and how it differed from the SI. In the end, it was possible to confirm that the level of availability of infrastructure services as provided through the SI will be higher than the native infrastructure after the migration if the proposed suggestions of this master thesis project are implemented successfully. These implementations also do not change the architecture of the SI in any major way. The end results of this thesis project were also highly appreciated by Ericsson and are now part of thedevelopment plan for next mobile core infrastructure solution at Ericsson.

Model-based design is a powerful design technique for embedded system development. The technique enables virtual prototyping to develop and debug controllers before touching real hardware. There are many tools available covering the distinct steps of the design cycle including modeling, simulation, and implementation. Unfortunately, none of them covers all three steps. This thesis proposes a formalism coupling the model and the implementation of a controller for equation-based simulation tools. The resulting formalism translates defined controller models to platform specific code using a defined set of syntax. A case study of a line-following robot has been developed to illustrate the feasibility of the approach. The prototype has been tested and evaluated using a sequence of test scenarios of increasing difficulty. The final experiments suggest that the behaviors of both modeled and generated controllers are similar. The thesis concludes that the approach of model-implementation coupling of controllers in the simplest form is feasible for equation-based tools. This allows it to conduct the whole model-based design cycle within a single environment.

No description available.