Mini solar power USB mobile charger using Buck-boost converter and MPPT

Help needed-Mini solar power USB mobile charger using Buck-boost converter and MPPT

Hi. I am designing a solar powered mini usb mobile charger as course project with switch mode power supply and maximum power point tracking. I am using a Flexible Ascent Solar 12V, 3W solar panel. Since, the project is for Switch Mode Power Supply(SMPS) course the charger must contain an SMPS with MPPT. I am assuming the maximum input voltage from the solar panel to be 9V and output voltage fixed at 5V. However, the MPPT will adjust the output current to supply maximum power to the load which in this case is the usb. The nominal value of the output current is 50mA. I am thinking of using a non-inverting buck-boost converter since it will be operational for both lower and higher than 5V at input. Also, I am planning to implement MPPT digitally using PIC18f4502. The problem I am facing is finding a Buck-boost converter IC with voltage specs as mentioned before with output current control and operating frequency range of 175-275 kHz.
I am considering a 0.75% variation in the output voltage of 5V since the usb can handle a variation from 4V to 6V. That is why I am considering my ADC step size to be 75 mV. Using this step size the minimum DPWM resolution comes out to 7 bits. Since, the PIC18f4520 can give a PWM resolution of 7 bits till 312.5 kHz with a 40 MHz oscillator I am hoping that using a buck-boost IC with operational frequency below 312.5 kHz will be able to work out. I am going to use Perturb duty cycle and Observe Uphill iterative MPPT algorithm. I am just worried that with 7 bits I will be able to generate 2^7 values for the duty cycle and I don't know what maximum duty cycle will I be able to generate in %age with this resolution?
I am attaching my block diagram of the project but the buck-boost IC in the diagram is not going to work since it adjusts the output voltage in the feedback loop and not the current. Also, I don't want to adjust the current in the feedback loop since it will become really difficult to close it. I looked at TI website and found two alternates lm5118 and lm25118 but i am not sure if they will serve my purpose. So, if someone can propose an alternate for the Buck-boost ic that will be really helpful. Also, any insight into the project will be highly appreciated.

Guys I would really appreciate some help here.

I would make it easier on myself if I were you. That solar panel has a Voc of 17.5V. The Vmp of that solar panel is probabaly around 13V under STC (1000W/m^2, 25degC) conditions. Vmp will not vary much with changes in light level (it is Imp that varies with irradiation); rather it is temperature dependent. Vmp will vary maybe 0.3%/degC, so not very much. The only time you will see a large drop in Vmp is if you shade a good portion of the panel. If it is dropping down below 5V, you are getting almost no power from the module anyway. So, I would say why bother with buck-boost and simplify your problem with a buck.